Despite numerous intervention efforts, risky drinking among adolescents and emerging adults remains prevalent, which is concerning given its association with other risk behaviors (e.g., drug use, drinking/driving) and health consequences (e.g., injury, overdose). The integration of alcohol interventions into popular social media platforms is a potentially exciting yet underutilized public health approach. Social media provides frequent interaction with online social networks, increasing exposure to peer influences. The most popular is Facebook, with 90% of adolescents and emerging adults having Facebook accounts that they access at least daily. Unlike traditional static website interventions, a key feature of social media is that content is dynamic, being constantly updated by users. In order to change alcohol use norms and behaviors, critical knowledge gaps in this area are how to capitalize on peer interaction in order to increase user-generated content, and identify intervention content that has the greatest diffusion. Thus, the proposed study will recruit adolescents and emerging adults (ages 16-24) using Facebook ads, and conduct online e-screening, enrolling 850 risky drinkers, who screen positive on the AUDIT-C, in a randomized controlled trial comparing three conditions: 8- week Social Media Intervention + Diffusion Incentives [i.e., gamification in which points earned for diffusion of participant's posts (responses by peer participants) result in monetary incentives]; 8-week Social Media Intervention Only, or 8-week attention placebo e-news control condition. Intervention conditions will involve access to secret Facebook group pages, separate by age group (16-20; 21-24), facilitated by peer e-health coaches, who will post dynamic content primarily focusing on alcohol use, but also addressing other drug use (illicit, non-medical prescription). Outcomes will be measured at 3, 6, and 12 months. The specific aims of the study are to: 1) develop and test the efficacy of intervention conditions compared to the control condition, in reducing risky drinking and alcohol related consequences; 2) compare active intervention conditions on participant engagement and alcohol-related outcomes; and, 3) examine how level of engagement in intervention conditions (i.e., Facebook metrics) and characteristics of intervention engagement (i.e., sentiment analysis) relate to alcohol-related outcomes. In addition, secondary aims are to examine: 1): the efficacy of the interventions, compared to the control condition, on reducing other drug use; 2) gender, age group (16-20; 21- 24), and baseline social network factors as moderators of outcome; and, 3) costs of the interventions to provide data to inform future public health interventions using social media. These innovative design features will provide the critical next step in public health efforts to reduce risky drinking, by capitalizing on social media to harness peer influences and deliver interventions using a familiar platform. Findings from this study could have enormous public health impact by altering the alcohol risk trajectories of adolescents and emerging adults, preventing health consequences as well as the development of alcohol use disorders.